- Blind mode tutorial
lichess.org
Donate

[Bug Report] Error parsing certain (not broken) PGN files/games

I'm trying to create a study from a PGN file made by Chessbase.

I tried both pasting and upload the file, but always got "Cannot parse moves: [26.41] failure: end of input expected" or similar errors, as well as an indicative on where the error occurred (even though it wasn't very helpful).

My PGN file contains 12 annotated games exported by Chessbase and had it's integrity checked by the program and then visually by me. No problems, encoding is UTF-8 (without BOM) as expected, there are no unclosed brackets or anything.

Chesstempo (https://www.chesstempo.com/pgn-viewer) can parse this file flawlessly. Lichess can correctly parse the rest of the file when I remove the game involved in the error (I had to remove games more than once in order to get it to be fully parsed).

I investigated whether notation commentary in plain text form (as in "white should have played 1. Bxc4 d5 2. Be2" < This is plain text in Chessbase) was causing it and also if parenthesis on plain text comments were breaking the PGN somehow, but by comparison, both things were present in other games from within the PGN with no errors.

After that, I ran out of ideas on how to further approach the problem.

I'm looking for confirmation if this is indeed a bug and, if a fix won't be possible soon for given reason, what exactly this problem is in order to circumvent it in my classes.

Any help is much appreciated, thank you!

Edit: stupid me forgot to attach the PGN, here it is: https://pastebin.com/XnxTcuQ0

I'm trying to create a study from a PGN file made by Chessbase. I tried both pasting and upload the file, but always got "Cannot parse moves: [26.41] failure: end of input expected" or similar errors, as well as an indicative on where the error occurred (even though it wasn't very helpful). My PGN file contains 12 annotated games exported by Chessbase and had it's integrity checked by the program and then visually by me. No problems, encoding is UTF-8 (without BOM) as expected, there are no unclosed brackets or anything. Chesstempo (https://www.chesstempo.com/pgn-viewer) can parse this file flawlessly. Lichess can correctly parse the rest of the file when I remove the game involved in the error (I had to remove games more than once in order to get it to be fully parsed). I investigated whether notation commentary in plain text form (as in "white should have played 1. Bxc4 d5 2. Be2" < This is plain text in Chessbase) was causing it and also if parenthesis on plain text comments were breaking the PGN somehow, but by comparison, both things were present in other games from within the PGN with no errors. After that, I ran out of ideas on how to further approach the problem. I'm looking for confirmation if this is indeed a bug and, if a fix won't be possible soon for given reason, what exactly this problem is in order to circumvent it in my classes. Any help is much appreciated, thank you! Edit: stupid me forgot to attach the PGN, here it is: https://pastebin.com/XnxTcuQ0

i would like to know why I was timed out for 1 hour on sat night,the only time ive gotten to play this week,i have costant connection problems on this site not to mention a lot of low caliber personalities,i would like a reply to sokon9969@gmail.com if yopu don't want me using your site anymore ill figure something else out thank you

i would like to know why I was timed out for 1 hour on sat night,the only time ive gotten to play this week,i have costant connection problems on this site not to mention a lot of low caliber personalities,i would like a reply to sokon9969@gmail.com if yopu don't want me using your site anymore ill figure something else out thank you

Adding to that, it was clarified that the error was being caused by null moves ("Z0" on the PGN or "--" on Chessbase), removing variations containing those is a valid workaround,

Adding to that, it was clarified that the error was being caused by null moves ("Z0" on the PGN or "--" on Chessbase), removing variations containing those is a valid workaround,

This topic has been archived and can no longer be replied to.